In this study, participants (N = 1500) were randomly assigned to take part in the experiment as part of the 1st of 2nd 750 participants. Within each batch of 750 participants, everyone played a virtual game of ‘cyberball’ where they imagined themselves playing a game of catch watching an avatar for themselves and two other players pass the ball back and forth. In the include condition, the avatars passed the ball to the participant as often as to the other avatar. In the ostracism condition, the avatars began by passing the ball to participants normally but then stopped throwing the ball to the participant, ‘playing’ amongst themselves. Afterwards, all participants were asked about much they trusted other people, how fairly other people treat them, and how helpful other people are. These three variables were averaged together into a generalized trust index. The prediction was that by being excluded, even by computer avatars, would lead to a decrease in generalized trust. Consistent with our pre-registration, we analyzed those who were in the 1st batch of 750 participants, then the 2nd batch, then aggregated them together for the complete N = 1500. There were no differences between the two 750s in terms of the magnitude of the ostracism effect (p > .77). For all analyses, we follow the originating lab and create a replication model using an OLS estimator and conditioning on the following demographic variables: sex, age, whether Hispanic, ethnicity (Black, White, Other), highest level of education, income, and U.S. census region, including a missingness indicator for each variable. We also provide average treatment effect (ATE) estimates in an additional analysis using robust standard errors to provide less model-reliant results. All reported means, standard deviations, and effect sizes represent marginal predicted means form the replication model. 1st 750 In the 1st 750 participants, we were able to confirm the effect that visualizing onesself in a cyberball game and being ostracized caused people to show less generalized trust (M = 3.124, SD = .891, n = 371) than if they had been included in the ball tossing game (M = 3.409, SD = .894, n = 379; b (719) = -.285, p < .001, d = -.32, 95%CI = -.463 to -.175; see Figure 1). The results remain the same using the ATE (b (742) = -.243, p < .001). 2nd 750 In the 2nd 750 participants, we found the same results. Visualizing oneself in a cyberball game and being ostracized caused people to show less generalized trust (M = 3.12, SD = . .889, n = 354) than if they had been included in the ball tossing game (M = 3.431, SD = .887, n = 394; b (722) = -.307, p < .001, d = -.343, 95%CI = -.487 to -.198; see Figure 2). The results remain the same using the ATE (b (743) = -.312, p < .001). Full 1500 Mirroring the results of its two constituent 750s, in the full 1500 participants, visualizing oneself in a cyberball game and being ostracized caused people to show less generalized trust (M = 3.121, SD = .885, n = 724) than if they had been included in the ball tossing game (M = 3.422, SD = .881, n = 723; b (1463) = -.299, p < .001, d = -.335, 95%CI = -.437 to -.233; see ). The results remain the same using the ATE (b (1487) = -.278, p < .001).